1. Field of the Invention
The present invention relates to content tagging and more particularly to taxonomy generation from tag data.
2. Description of the Related Art
Content browsing refers to the retrieval and presentation of electronic content in a browser client. Content generally can include electronic documents, messages, audio, audiovisual and video materials and imagery. Most commonly, content can be stored in a server environment and published for access by content consumers over a computer communications network such as the global Internet. Content consumers, in turn, can retrieve content over the network by reference to a network address for the content. Once retrieved, the content can be presented in a browser client including not only conventional visual browsers such as the venerable Web browser, but also in document viewers and editors, alternative browsers such as those deployed in pervasive devices and those supporting different modes of presentation such as the audible presentation of material.
Given the vast amount of content published for accessibility over the Internet, modern content browsers provide a mechanism for ably retrieving previously accessed content. Known as a “bookmark” or “favorite” (collectively referred to as a bookmark), end users can record content of interest as a bookmark. Subsequently, end users can access a list of bookmarks in order to recall the content of interest without being compelled to recall from memory the precise uniform resource indicator (URI) for the content of interest. While bookmarking reflects the personal experience of individual users, social bookmarking provides a foundation for users within a social group to store, organize, share and search the bookmarks collectively established by the users within the social group.
In the operation of a social bookmarking system, users save links to memorable content. Unlike traditional bookmarks however, in a social bookmarking system the links subsequently can be published for public inspection and use so as to provide a communal repository of bookmarks. Consequently, groups of the users can access the links encapsulated within respective social bookmarks, though the groups of users in fact may never have viewed the associated content—a prerequisite for a traditional bookmark.
Social bookmarking services often encourage users in a social network to annotate bookmarks with meta-information referred to as “tags” rather than merely storing bookmarks in a traditional file hierarchy. As such, users processing tags for a social bookmark can view the social bookmark for content along with the tag pertaining to the bookmark, for instance a number of users having bookmarked the content. Further, some social bookmarking services infer clusters of bookmarks from the relationship of corresponding tags. Finally, many social bookmarking services provide subscription based feeds for lists of bookmarks, including lists organized by tags. Consequently, subscribers can become aware of new bookmarks as the bookmarks are saved, shared, and tagged by other users.
Tags have been extended from an social bookmarks referencing content, to the direct referencing of content absent a social bookmark. Consequently, in the circumstance of referencing underlying content, tags reflect a “bottom-up” type of classification for the content, compared to the hierarchical structure of content such as in a file system or index, which are “top-down” in nature. In a traditional hierarchical system, also known as a taxonomy, the designer of the taxonomy establishes a limited number of terms to use for classification, resulting in only one correct way to classify each item of content. In a tagging system, however, there are an unlimited number of ways to classify an item, and there is no “wrong” choice. Instead of belonging to one category, an item of content may have been tagged with several different tags.